The purpose of this research is to develop methods of assessing the information needs and information seeking behavior of primary care practitioners, in particular rural practitioners. Valid and reliable assessment methods are needed (l) to inform the design of information systems for primary care practitioners; and (2) to evaluate these systems by measuring their impact on practitioners' information needs and information seeking behavior. The specific aims are: A. Expand current understanding by assessing the information needs and information seeking of a broad spectrum of practitioners. The PI will combine ethnographic observation with interview methods developed in earlier work to assess the information needs and information seeking of A.l. Non-volunteering primary care physicians, those who have not responded to surveys in the past, but represent the majority of those in practice; and A.2. Primary care nurse practitioners and physician assistants, who play an increasingly important role in delivery of primary care, especially in rural areas, but whose information needs and information seeking behavior have not been examined. B. Develop a taxonomy of information needs. No taxonomy of information needs now exists. Many studies have focused on the questions that physicians pursue. But seventy percent of physicians' clinical questions are never pursued, and these questions have not been adequately examined. A taxonomy of information needs that distinguishes needs that are being met from those that are not is necessary to improve understanding of information resources use, and to measure the impact of new information systems. C.Improve existing methodology for assessing information needs and outcomes. Standard metrics do not exist for most variables relevant to information needs assessment. Valid, reliable, and affordable methods are needed for comparative studies of information system impact. This research will C.1. Compare and calibrate alternate information needs assessment methodologies, including ethnographic observation, semi-structured interviews, and microcassette recorder logs; C.2. Examine the usefulness of new measures of the outcomes of information seeking, including the quality of the information obtained and it's effect on practitioner uncertainty about patient care. D. Apply these methods to evaluate new information systems by measuring their impact on the information needs and information seeking behavior of primary care practitioners. The multi-method approach and new measures developed as described will be used to assess new systems: D.l. Evaluation of a statewide information system incorporating gopher- based internet access on the information management of primary care practitioners in rural and nonrural practices; D.2. Evaluation of an academic information system employing Wide Area Information Server technology on the information management of physicians on a medical center campus.